Skip to main content
Device Global Header

Subscribe
or Renew

Advanced Search

Original Article

Effect of Eicosapentaenoic and Docosahexaenoic Acids on Blood Pressure in Hypertension — A Population-Based Intervention Trial from the Tromsø Study

List of authors.
  • Kaare H. Bønaa, M.D.,
  • Kristian S. Bjerve, M.D.,
  • Bjørn Straume, M.D.,
  • Inger T. Gram, M.D.,
  • and Dag Thelle, M.D.

Abstract

Studies of whether polyunsaturated fatty acids in fish oil — in particular, eicosapentaenoic and docosahexaenoic acids — lower blood pressure have varied in design and results. We conducted a population-based, randomized, 10-week dietary-supplementation trial in which the effects of 6 g per day of 85 percent eicosapentaenoic and docosahexaenoic acids were compared with those of 6 g per day of corn oil in 156 men and women with previously untreated stable, mild essential hypertension.

The mean systolic blood pressure fell by 4.6 mm Hg (P = 0.002), and diastolic pressure by 3.0 mm Hg (P = 0.0002) in the group receiving fish oil; there was no significant change in the group receiving corn oil. The differences between the groups remained significant for both systolic (6.4 mm Hg; P = 0.0025) and diastolic (2.8 mm Hg; P = 0.029) pressure after control for anthropometric, lifestyle, and dietary variables. The decreases in blood pressure were larger as concentrations of plasma phospholipid n—3 fatty acids increased (P = 0.027). Dietary supplementation with fish oil did not change mean blood pressure in the subjects who ate fish three or more times a week as part of their usual diet, or in those who had a base-line concentration of plasma phospholipid n—3 fatty acids above 175.1 mg per liter.

We conclude that eicosapentaenoic and docosahexaenoic acids reduce blood pressure in essential hypertension, depending on increases in plasma phospholipid n—3 fatty acids. (N Engl J Med 1990; 322:795–801.)

Introduction

IN 1951 Ehrström1 noted that "arterial hypertension of the malignant type has never been diagnosed in North Greenland and essential hypertension is rare." The results of more recent studies of blood pressure in Eskimo populations are conflicting.2 3 4 5 The apparent low death rate from coronary heart disease among Eskimos6 has focused interest on the potential beneficial effects of the polyunsaturated fatty acids in fish oil on blood lipid and lipoprotein levels,7 platelet aggregability,8 and antiinflammatory or immunologic effects.9

Fish oils have lowered blood pressure in normotensive10 11 12 13 14 and hypertensive15 16 17 18 subjects in some but not all intervention trials.19 , 20 The extent to which these reports can be generalized is constrained by limitations in study design, particularly in cases in which large doses of oil were given and the dietary changes were not documented or controlled.21

We therefore conducted a population-based, randomized, double-blind study comparing the effects of fish oil and corn oil to test the hypothesis that supplementing the usual Western diet with eicosapentaenoic and docosahexaenoic acids would reduce blood pressure in hypertension. We also monitored diet and assessed levels of plasma phospholipid fatty acids to determine the relation between diet, fatty acids, and blood pressure.

Methods

Subjects

In 1986—1987, 21,826 subjects — 81.3 percent of the men 20 to 61 years old and the women 20 to 56 years old living in the municipality of Tromsø, Norway — participated in a health survey. All completed a questionnaire about previous cardiovascular disease, diet, and smoking habits, and their weight, height, blood pressure, and nonfasting serum lipid concentrations were measured. All 299 subjects between the ages of 34 and 60 who were not receiving antihypertensive treatment, whose diastolic blood pressure was above 97 mm Hg, and whose systolic blood pressure was below 210 mm Hg were asked to undergo clinical examination that included a complete history, physical examination, laboratory tests, and electrocardiography. Among the 236 subjects who underwent the examination, 201 who had no cardiovascular disease, bleeding disorder, diabetes mellitus, disabling chronic disease, psychopathologic disease, alcoholism, or extreme obesity (body-mass index [weight in kilograms divided by the square of the height in meters] >32) were recruited into this study. The study was approved by the regional board of research ethics, and each subject gave written informed consent. The study began with a six-month run-in period during which the subjects were asked to continue their usual diet and living habits and during which their blood pressure was measured on three occasions.

Table 1. Table 1. Composition of the Fish-Oil and Corn-Oil Supplements.

For entry into the intervention phase of the trial, the mean diastolic pressure during the run-in period had to be between 85 and 110 mm Hg, and the mean systolic blood pressure below 180 mm Hg. Among the 201 subjects, 21 had blood-pressure values that did not meet these criteria, 13 dropped out for personal reasons, and 10 were found to have one or more of the disorders listed above. Thus, 157 subjects entered the 10-week intervention trial. They were randomly assigned to receive fish oil or corn oil after stratification according to sex. The fish-oil group was given 6 g of 85 percent eicosapentaenoic and docosahexaenoic acids (ethyl ester K85, Norsk Hydro, Oslo, Norway) per day. This increased their average intake of these acids from 1.0 to 6.1 g per day. The corn-oil group received 6 g of corn oil per day, which increased their dietary intake of linoleic acid from 8.9 to 12.3 g per day. The oils (Table 1) were given in indistinguishable soft gelatin capsules, each containing 1 g of fatty acids. Either supplement provided approximately 2 percent of the mean daily energy intake. The subjects were again asked at the start of the intervention phase not to change their diets or living habits.

After 5 and 10 weeks of supplementation with fish oil or corn oil, the participants were examined between 8 a.m. and 1 p.m. after an overnight fast. They were asked not to smoke during the two hours before the examination. Compliance was assessed by counting leftover capsules, and the subjects were questioned about side effects and intercurrent diseases.

Clinical and Laboratory Measurements

Blood pressure was recorded before blood sampling by a specially trained secretary using an automatic device (Dinamap, Critikon, Tampa). The instrument measured blood pressure by the oscillometric method22 and calculated the mean arterial pressure automatically by dividing the area under the pressure wave form by the time during which the area was measured.22 After the subjects had been seated for two minutes, three recordings were made at two-minute intervals. The mean of the last two measurements was used. Then, two measurements were made after the subjects had been standing for two minutes, and the mean values were calculated. The recorder was calibrated with the same mercury sphygmomanometer at regular intervals. Body weight was measured with an electronic scale, with the subjects wearing light clothing.

Each subject's average intake of nutrients was estimated by a certified clinical nutritionist using 24-hour dietary-recall records. Interviews were conducted five weeks before the start of the intervention trial and at its end.23 Dietary constituents were calculated from standard Norwegian food tables.24

Blood samples were obtained at the beginning and end of the 10-week intervention period. Plasma and serum were stored at — 80°C until the study was completed, and they were analyzed before the randomization code was broken. Serum high-density lipoprotein (HDL) cholesterol was measured after the precipitation of serum with phosphotungstic acid and magnesium chloride.25 Serum total cholesterol and triglyceride levels were analyzed by enzymatic colorimetric methods with commercial kits (CHOD-PAP for cholesterol and GPO-PAP for triglycerides; Boehringer–Mannheim, Mannheim, Federal Republic of Germany). Both samples from each patient were analyzed at the same time. Plasma lipids were extracted with n-butanol (butyl alcohol).26 The phospholipids were isolated from these extracts by column chromatography after diheptadecanoyl-glycerophosphocholine and butylated hydroxytoluene (Sigma Chemical, Pool, United Kingdom) had been added as an internal standard and antioxidant, respectively. The phospholipids were transmethylated and quantitated by gas—liquid chromatography.27 A normal human serum sample was included as a control to monitor analytical performance. The day-to-day coefficients of variation (n = 55) for 20:4n—6, 20:5n—3, and 22:6n—3 fatty acids were 3.8, 3.7, and 4.7 percent, respectively. The results were quantitated as milligrams of phospholipid fatty acid per liter of plasma.

Bleeding time was measured by the Simplate-II method (General Diagnostics, Organon Teknika, Turnhout, Belgium), platelet counts and plasma fibrinogen levels by standard laboratory methods, and serum alpha-tocopherol levels by high-performance liquid chromatography.28

Statistical Analysis

The primary end point was a change in blood pressure between the beginning (base line) and the end of the 10-week intervention trial. The study was designed to detect a difference of 3.5 mm Hg in diastolic blood pressure at a two-sided level of significance of 0.05, with a power of 0.90. We analyzed the data in three ways. First, the changes in blood pressure in the fish-oil and corn-oil groups were compared. Second, we compared changes in blood pressure within and between strata divided according to the usual consumption of fish, using results obtained during the run-in period, and strata divided according to concentrations of plasma phospholipid n—3 fatty acids. In these analyses we used the difference between the beginning blood pressure and the arithmetic mean of the blood pressures measured after 5 and 10 weeks. Mean arterial pressure was used in subgroup analyses. Third, the association between the change in blood pressure and the change in concentrations of plasma phospholipid fatty acids was examined. The data were analyzed for statistical significance by t-tests and by chi-square statistics when appropriate. Analysis of covariance was used for tests of interaction and for adjustments. The association between the change in blood pressure and the change in concentrations of plasma phospholipid n—3 fatty acids was tested by computing simple and partial Pearson's correlation coefficients (adjusted for age, sex, change in weight, energy intake, and concentrations of plasma phospholipid saturated and monounsaturated fatty acids) with pooled data from both supplementation groups. Linear trend was assessed by analysis of variance with contrasts (−3,−1,1,3) between means of categories. The SAS software package29 was used. All results are given as means ±SD. P values lower than 0.05 were considered to indicate significance.

Results

Table 2. Table 2. Characteristics of the Subjects with Hypertension in the Fish-Oil and Corn-Oil Groups.* Table 3. Table 3. Intake of Nutrients and Body Weight before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*

Table 2 shows that the groups assigned to receive fish oil and corn oil were well balanced at base line. All 157 subjects completed the study, but the results from 1 man were excluded from our analysis because he had a myocardial infarction and was then treated with cardiac drugs; our results are therefore based on the data from 156 subjects. The reported side effects were mild or moderate abdominal problems, such as belching, self-limited diarrhea, and obstipation. Their frequencies were similar in the two groups after 5 weeks, but after 10 weeks 19 subjects in the corn-oil group and 9 in the fish-oil group reported such problems. The subjects receiving corn oil and fish oil took 94.2 and 90.4 percent, respectively, of the prescribed number of capsules. There were no dietary changes during the trial in the fish-oil group, whereas in the corn-oil group there was a small reduction in dietary carbohydrate and dietary n—3 fatty acids (Table 3). The mean body weight increased 0.70 kg in the fish-oil group and 0.56 kg in the corn-oil group (Table 3). Serum alpha-tocopherol levels did not change in the fish-oil group, but increased slightly in the corn-oil group. Bleeding time, platelet count, and plasma fibrinogen levels did not change significantly in either group.

There was no change in serum total cholesterol levels in either group. Serum HDL cholesterol values increased from 1.34 to 1.41 mmol per liter (5.2 percent; P = 0.020) among those receiving fish oil and from 1.33 to 1.40 mmol per liter (5.3 percent; P = 0.002) among those receiving corn oil. The change was not significantly different (P = 0.69) between groups. In the fish-oil group, serum triglyceride levels decreased from 1.48 to 1.18 mmol per liter (20.3 percent; P = 0.0007), whereas in the corn-oil group serum triglyceride levels did not change.

Plasma Concentrations of Phospholipid Fatty Acids and Consumption of Fish

Table 4. Table 4. Plasma Phospholipid Fatty Acid Concentrations before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*

The mean plasma concentration of phospholipid n—3 fatty acids increased and that of phospholipid n—6 fatty acids decreased in the group receiving fish oil (Table 4), increasing the ratio of eicosapentaenoic to arachidonic acid from 0.42 to 1.22 (P<0.0001). In the corn-oil group the mean plasma concentration of phospholipid n—6 fatty acids increased, and that of phospholipid n—3 fatty acids decreased. There were no significant changes in plasma levels of phospholipid total or saturated fatty acids in either group. In a subgroup of 22 randomly selected men from both groups, the fatty acid distribution in platelet phospholipids was measured (data not shown). The Pearson's correlation coefficient between the changes in the concentrations of eicosapentaenoic acid in platelets and in plasma phospholipids was 0.91 (P<0.0001), indicating that the changes in the fatty acid composition of biologic membranes during the study were similar to those in plasma levels of phospholipid fatty acids.

Exactly 50 percent of the subjects reported consuming three or more fish dishes a week on their usual diet. Among them were 44 in the corn-oil group and 34 in the fish-oil group. When compared with these subjects, those who consumed fewer than three fish dishes a week had lower base-line plasma concentrations of phospholipid n—3 fatty acids (169.2±47.5 vs. 199.4±65.3 mg per liter; P = 0.0014) and higher concentrations of n—6 fatty acids (450.7±75.3 vs. 414.0±77.8 mg per liter; P = 0.0038). The plasma concentrations of phospholipid saturated and monounsaturated fatty acids were similar. Within the fish-oil group, those with a low intake of fish had a larger mean increase in plasma levels of phospholipid n—3 fatty acids, as compared with those with a high intake of fish (71.9 vs. 58.7 percent), and a larger mean decrease in plasma concentrations of phospholipid linoleic acid (19.7 vs. 17.5 percent) during supplementation.

Base-Line Blood Pressure

Table 5. Table 5. Comparison of Changes in Blood Pressure before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*

The subjects in both groups with a high intake of fish had lower base-line mean arterial pressure than those with a low intake of fish (111.5±10.7 vs. 114.8±8.8 mm Hg; P = 0.043) when the values were adjusted for age, sex, and body-mass index. The determinants of base-line mean arterial pressure were analyzed in a multiple regression analysis that included base-line anthropometric, lifestyle, and dietary variables and base-line plasma levels of phospholipid fatty acids. Among these fatty acids, only eicosapentaenoic acid (b = −0.063, P = 0.018) was included in the model (R2 = 0.12, P = 0.004) as a significant independent predictor of blood pressure. The model predicted a 4.4 mm Hg reduction in mean arterial pressure if plasma levels of phospholipid eicosapentaenoic acid increased by 69.6 mg per liter (the mean increase in the fish-oil group during the intervention period; Table 4), which is similar to the measured reduction in mean pressure during the period (Table 5).

Change in Blood Pressure during Oil Supplementation

The blood-pressure values during the run-in period and at base line were similar, indicating that a stable level had been reached (Table 2). The subjects receiving fish oil had a significant reduction in systolic (4.6 mm Hg; P = 0.002), diastolic (3.0 mm Hg; P = 0.0002), and mean (3.5 mm Hg; P = 0.002) blood pressure in the sitting position and in systolic, diastolic, and mean blood pressure in the standing position, as compared with base line, at the end of the 10-week intervention period. In contrast, the subjects receiving corn oil had no significant changes in any of these measurements (Table 5). The differences between the two groups were significant for all mean pressure values and all values measured in the sitting and standing positions. No significant interaction was found with sex, age, or body weight. After adjustment for age, sex, smoking habits, usual consumption of fish, change in body weight, intake of energy, dietary fiber, sodium, and calcium, consumption of alcohol, and change in plasma concentrations of phospholipid saturated and monounsaturated fatty acids, the estimated decrease in the fish-oil group was 5.0 mm Hg (95 percent confidence limits, 2.2 and 7.8) for systolic pressure and 3.1 mm Hg (95 percent confidence limits, 1.4 and 4.8) for diastolic pressure. The adjusted differences between the two groups remained significant for both systolic (6.4 mm Hg; P = 0.0025) and diastolic (2.8 mm Hg; P = 0.029) pressure. In the fish-oil group, 30.8, 16.7, and 20.5 percent of the subjects had reductions in systolic pressure of more than 10 mm Hg, between 5.1 and 10 mm Hg, and between 0.1 and 5 mm Hg, respectively, and 32 percent had no change or an increase. The corresponding figures in the corn-oil group were 11.5, 24.4, 17.9, and 46.2 percent (P = 0.018).

Figure 1. Figure 1. Changes in Mean Arterial Pressure during the Study, According to Group and Number of Fish Dishes per Week in the Usual Diet.

Each point represents the group mean adjusted for age, sex, smoking (yes or no), and changes in body weight from week 0 to week 10; standard errors of the mean ranged from 1.4 to 1.8. The differences between values at week 0 and the arithmetic means of the measurements at week 5 and week 10 were tested by analysis of variance.

The change in blood pressure differed significantly (P<0.0001) between the subjects with a low intake of fish in the fish-oil group (■—■) and in the corn-oil group (■— —■), but not (P = 0.28) between the subjects with a high intake of fish in the fish-oil group (▲—▲) and in the corn-oil group (▲— —). The change in blood pressure differed significantly between the subjects with high and low intakes of fish within both the fish-oil (P = 0.0008) and the corn-oil (P = 0.048) groups.

Figure 2. Figure 2. Changes in Mean Arterial Pressure during the Study, According to Group and Base-Line Plasma Phospholipid n—3 Fatty Acid Concentrations.

The subgroups are based on the median base-line plasma concentration of phospholipid n—3 fatty acids (the sum of alpha-linolenic, eicosatetraenoic, eicosapentaenoic, docosapentaenoic, and docosahexaenoic acid levels), which was 172.6 mg per liter in the corn-oil group and 175.1 mg per liter in the fish-oil group. Each point represents a group mean adjusted for age, sex, smoking (yes or no), and change in body weight from week 0 to week 10; standard errors of the mean ranged from 1.6 to 1.8. The differences between the value at week 0 and the arithmetic mean of the measurements at week 5 and week 10 were tested by analysis of variance. The values for two subjects in the fish-oil group and three in the corn-oil group were not available, and those subjects were not included in these analyses. Squares represent subjects below the group median, and triangles those above. Solid lines represent the fish-oil group, and broken lines the corn-oil group.

The change in blood pressure differed significantly between the subjects below the group median in the fish-oil group and those below the group median in the corn-oil group (P = 0.0024), as well as between the two strata within the fish-oil group (P = 0.016).

The change in blood pressure was related to the amount of fish usually consumed (Fig. 1). The 44 subjects in the fish-oil group with a low intake of fish had a decrease in mean arterial pressure (6.1 mm Hg; 95 percent confidence limits, 8.8 and 3.4; P<0.0001), whereas mean arterial pressure increased in the 34 subjects in the corn-oil group with a low intake of fish (3.1 mm Hg; 95 percent confidence limits, 0.1 and 6.1; P = 0.036). Among the subjects who ate more fish, supplementation with fish oil or corn oil did not change the mean pressure significantly. The validity of these findings was supported by the similar results obtained when the participants were grouped according to base-line plasma concentrations of phospholipid n—3 fatty acids (Fig. 2). A multiple regression analysis was performed within the fish-oil group, with change in blood pressure as the dependent variable and the amount of fish usually consumed, age, weight, sex, smoking habits, and base-line plasma concentrations of phospholipid saturated and monounsaturated fatty acids as independent variables. In this model (R2 = 0.28, P = 0.0029) fish consumption alone (partial R2 = 0.19, P = 0.0001) accounted for 67.9 percent of the explained variance.

Figure 3. Figure 3. Changes in Mean Arterial Pressure during the Study, According to Changes in Plasma Phospholipid n—3 Fatty Acid Concentrations in the Fish-Oil Group.

The values are the mean (+SEM) differences between base-line values and the arithmetic mean of measurements at week 5 and week 10, adjusted for age, sex, smoking (yes or no), and change in body weight (P = 0.027 for linear trend by analysis of variance). The values for two subjects were unavailable and those subjects were not included in these analyses. The changes in plasma phospholipid n—3 fatty acid levels were measured as the percent change (by quartiles) in the sum of alpha-linolenic, eicosatetraenoic, eicosapentaenoic, docosapentaenoic, and docosahexaenoic acids between base line and 10 weeks.

There was a significant inverse relation between the change in systolic blood pressure and the change in the plasma concentration of phospholipid eicosapentaenoic acid in the entire group of 156 subjects analyzed (simple and partial Pearson's correlation coefficients, −0.27 [P = 0.0007] and −0.32 [P = 0.0003], respectively). The corresponding figures for the plasma level of phospholipid docosahexaenoic acid were −0.20 (P = 0.014) and −0.29 (P = 0.008). The trends for diastolic blood pressure were similar. In the fish-oil group the change in mean blood pressure was significantly (P = 0.027) related to the change in plasma levels of phospholipid n—3 fatty acids (Fig. 3).

Discussion

The interactions between dietary nutrients and blood pressure are complex yet highly relevant to contemporary attempts to reduce the risk of cardiovascular disease. The results of this population-based intervention study demonstrate that polyunsaturated fatty acids in fish oil lower blood pressure in subjects with hypertension. Several lines of evidence support a causal relation between the dietary intake of fish oil and blood pressure. First, the study hypothesis was confirmed in a randomized trial. Second, the change in blood pressure was related to the change in plasma concentrations of phospholipid n—3 fatty acids. Third, there was a significant association between plasma concentrations of phospholipid eicosapentaenoic acid and blood pressure. Fourth, the frequency of fish consumption predicted both base-line blood pressure and the response to supplementation.

The biologically active component of fish oil that reduces blood pressure is not known with certainty. We found that a mixture of eicosapentaenoic and docosahexaenoic acids lowered blood pressure. Although there was an association between the change in blood pressure and that in plasma concentrations of both phospholipid docosahexaenoic acid and phospholipid eicosapentaenoic acid, base-line blood pressure was related only to the latter. Our results thus indicate that eicosapentaenoic acid is more important in lowering blood pressure.

Our results further suggest that n—3 polyunsaturated fatty acids supplied as ethyl esters have a lowering effect on blood pressure that is quantitatively similar to the effect of those supplied as part of the usual diet. It has recently been reported that eicosapentaenoic acid ethyl ester is absorbed to the same extent as the triglyceride form.30 The apparently poor absorption reported in single-dose experiments31 , 32 probably reflects differences in the hydrolytic rate and tissue distribution of the ethyl and glycerol esters.

There was a slight increase in body weight (<1 percent) in both groups during the study. Weight gain has been reported in both dietary-supplementation and dietary-replacement studies using fish oil.9 , 19 It is not likely that the small amount of extra energy supplied by the supplement used in this study caused the increase in weight; other possible explanations include changes in the absorption or utilization of other nutrients and factors outside the study, such as seasonal variation.

Previous intervention studies have demonstrated that fish oils have hypotensive effects, in both normotensive10 11 12 13 14 and hypertensive15 16 17 18 subjects. In one open study in nonsmoking white men with hypertension,15 1.8 g of eicosapentaenoic acid per day did not lower blood pressure, whereas 9 g per day lowered systolic pressure by 6.5 mm Hg and diastolic pressure by 4.4 mm Hg. We reanalyzed our results using the same15 inclusion criteria (including only the men who consumed fewer than three fish dishes a week) and found that systolic pressure decreased by 11.2 mm Hg (P = 0.028) and diastolic pressure by 5.6 mm Hg (P = 0.013) in 9 men receiving fish oil, as compared with no change in 11 similar men receiving corn oil. Although the results of this subgroup analysis should be interpreted cautiously, they suggest that little additional lowering effect can be expected from increasing the dose of eicosapentaenoic acid from 3.3 g to 9 g per day. Furthermore, 1.8 g per day may be insufficient to achieve any effect at all, although this may depend on the composition of the usual diet.

The efficacy of fish-oil supplements in reducing blood pressure may vary in different demographic and biochemical subgroups of patients with hypertension. We found that 32 percent of the subjects receiving fish oil had no reduction in blood pressure despite an increased intake of eicosapentaenoic acid. As shown in Figures 1, 2, and 3, part of the variation in blood-pressure response was related to the specific dietary and biochemical characteristics of the subjects. Those following a diet lower in fish and those with low plasma concentrations of phospholipid n—3 fatty acids were particularly likely to benefit from fish-oil supplements. To study further potential differences between the subjects receiving fish oil who responded and those who did not, we compared those who had a decrease in systolic blood pressure of more than 10 mm Hg (n = 24) with those who had no change or an increase in blood pressure (n = 25). The subjects who responded to supplementation were less likely to eat fatty fish (P = 0.024), had higher base-line plasma concentrations of phospholipid monounsaturated (P = 0.041) and saturated (P = 0.058) fatty acids, and had a larger reduction in plasma levels of phospholipid n—6 fatty acids during the study (P = 0.059). We found no relation between the dietary intake of sodium, calcium, or fiber and the response to fish-oil supplements. This may imply that the amount or quality of dietary fat (or both) modifies the lowering effect of fish oil on blood pressure. It is tempting to speculate that a low ratio of n—3 to n—6 fatty acids in biologic membranes is associated with increased susceptibility to hypertension, and that blood pressure may be reduced in such persons by raising the n—3:n—6 ratio.

The effect of n—6 polyunsaturated fatty acids on blood pressure is controversial. Linoleic acid has been reported to decrease33 as well as to increase34 blood pressure, and to have no effect.35 Our study suggests that linoleic acid may increase blood pressure in patients with a low intake of n—3 fatty acids, but that it has little effect in those with a high intake of n—3 fatty acids (Fig. 1).

We have demonstrated that dietary enrichment with 6 g per day of 85 percent eicosapentaenoic and docosahexaenoic acids can lower blood pressure in subjects with hypertension. The blood-pressure response was related to change in plasma levels of phospholipid n—3 fatty acids. This relation implies that in addition to its potential benefits through other physiologic mechanisms,36 , 37 fish oil may be helpful in treating hypertension.

Funding and Disclosures

Supported by the Norwegian Research Council for Science and the Humanities, the Norwegian Fishermen Sales Organization, and Norsk Hydro; the screening was carried out in cooperation with the National Health Screening Service, Oslo, Norway.

We are indebted to our study participants for their cooperation; to Mrs. Beate Packalen, study secretary; to Mrs. Åse Borgejordet, who organized the dietary recall; to Mrs. Merethe Mack and Mrs. Birgit Svensson for technical assistance; to Dr. Ole C. Ingebrigtsen, head of the department of clinical chemistry at the university teaching hospital, Tromsø; and to Professor Christian Drevon for advice and measurements of alpha-tocopherol levels.

Author Affiliations

From the Institute of Community Medicine, University of Tromsø, Tromsø (K.H.B., B.S., I.T.G., D.T.), and the Department of Clinical Chemistry, Regional Hospital, University of Trondheim, Trondheim (K.S.B.), both in Norway. Address reprint requests to Dr. Bønaa at the Institute of Community Medicine, University of Tromsø, P.O. Box 417, N-9001 Tromsø, Norway.

References (37)

  1. 1. Ehrström MC. . Medical studies in North Greenland 1948–1949. VI. Blood pressure. Hypertension and arteriosclerosis in relation to food and mode of living . Acta Med Scand 1951; 140:416–22.

  2. 2. Bjerager P, Kromann N, Thygesen K, Harvald B. . Blood pressure in Greenland Eskimos . Ugeskr Læger 1980; 142:2278–80.

  3. 3. Simper LB. . Blood pressures in polar Eskimos . Ugeskr Læger 1976; 138:1757–8.

  4. 4. Bang HO, Dyerberg J. . Arktiske blodtryk . Ugeskr Læger 1976; 138:2503.

  5. 5. Jørgensen KA, Høj Nielsen A, Dyerberg J. . Hemostatic factors and renin in Greenland Eskimos on a high eicosapentaenoic acid intake: results of the Fifth Umanak Expedition . Acta Med Scand 1986; 219:473–9.

  6. 6. Kromann N, Green A. . Epidemiological studies in the Upernavik district, Greenland: incidence of some chronic diseases 1950–1974 . Acta Med Scand 1980; 208:401–6.

  7. 7. Phillipson BE, Rothrock DW, Connor WE, Harris WS, Illingworth DR. . Reduction of plasma lipids, lipoproteins, and apoproteins by dietary fish oils in patients with hypertriglyceridemia . N Engl J Med 1985; 312:1210–6.

  8. 8. Dyerberg J, Bang HO, Stoffersen E, Moneada S, Vane JR. . Eicosapentaenoic acid and prevention of thrombosis and atherosclerosis? Lancet 1978; 2:117–9.

  9. 9. Lee TH, Hoover RL, Williams JD, et al. . Effect of dietary enrichment with eicosapentaenoic and docosahexaenoic acids on in vitro neutrophil and monocyte leukotriene generation and neutrophil function . N Engl J Med 1985; 312:1217–24.

  10. 10. Singer P, Jaeger W, Wirth M, et al. . Lipid and blood pressure-lowering effect of mackerel diet in man . Atherosclerosis 1983; 49:99–108.

  11. 11. Sanders TAB, Vickers M, Haines AP. . Effect on blood lipids and haemostasis of a supplement of cod-liver oil, rich in eicosapentaenoic and docosahexaenoic acids, in healthy young men . Clin Sci 1981; 61:317–24.

  12. 12. Lorenz R, Spengler U, Fischer S. Duhm J, Weber PC. . Platelet function, thromboxane formation and blood pressure control during supplementation of the Western diet with cod liver oil . Circulation 1983; 67:504–11.

  13. 13. Mortensen JZ, Schmidt EB, Nielsen AH, Dyerberg J. . The effect of n—6 and n—3 polyunsaturated fatty acids on hemostasis, blood lipids and blood pressure . Thromb Haemost 1983; 50:543–6.

  14. 14. Rogers S, James KS, Butland BK, Etherington MD, O'Brian JR, Jones JG. . Effects of a fish oil supplement on serum lipids, blood pressure, bleeding time, haemostatic and rheological variables: a double blind randomised controlled trial in healthy volunteers . Atherosclerosis 1987; 63:137–43.

  15. 15. Knapp HR, FitzGerald GA. . The antihypertensive effects of fish oil: a controlled study of polyunsaturated fatty acid supplements in essential hypertension . N Engl J Med 1989; 320:1037–43.

  16. 16. Singer P, Berger I, Lück K, Taube C, Naumann E, Gödicke W. . Long-term effect of mackerel diet on blood pressure, serum lipids and thromboxane formation in patients with mild essential hypertension . Atherosclerosis 1986; 62:259–65.

  17. 17. Singer P, Wirth M, Voight S, et al. . Blood pressure- and lipid-lowering effect of mackerel and herring diet in patients with mild essential hypertension . Atherosclerosis 1985; 56:223–35.

  18. 18. Norris PG, Jones CJH, Weston MJ. . Effect of dietary supplementation with fish oil on systolic blood pressure in mild essential hypertension . Br Med J 1986; 293:104–5

  19. 19. von Houwelingen R, Nordøy A, van der Beek E, Houtsmuller U, de Metz M, Hornstra G. . Effect of a moderate fish intake on blood pressure, bleeding time, hematology, and clinical chemistry in healthy males . Am J Clin Nutr 1987; 46:424–36.

  20. 20. Demke DM, Peters GR, Linet OI, Metzler CM, Klott KA. . Effects of a fish oil concentrate in patients with hypercholesterolemia . Atherosclerosis 1988; 70:73–80.

  21. 21. Bønaa K. . Epidemiological and intervention studies on the effect of marine polyunsaturated fatty acids on blood pressure . J Intern Med Suppl 1989; 225:105–10.

  22. 22. Meldrum SJ. . The principles underlying Dinamap — a microprocessor based instrument for the automatic determination of mean arterial pressure . J Med Eng Technol 1978; 2:243–4.

  23. 23. Callmer E, Hagman U, Haraldsdottir J, Løken EB, Seppänen R, Trygg K. . Standardisation of 24-hour recall . Vår Føda 1986; 38:Suppl 4:250–71.

  24. 24. Statens ernæringsrÅd. Matvaretabell (Norwegian food table). Oslo, Norway: Landsforeningen for kosthold og helse, 1984.

  25. 25. Lopes-Virella MF, Stone P, Ellis S, Colwell JA. . Cholesterol determination in high-density lipoproteins separated by three different methods . Clin Chem 1977; 23:882–4.

  26. 26. Bjerve KS, Daae LNW, Bremer J. . The selective loss of lysophospholipids in some commonly used lipid-extraction procedures . Anal Biochem 1974; 58:238–45.

  27. 27. Bjerve KS, Fischer S, Alme K. . Alpha-linolenic acid deficiency in man: effect of ethyl linolenate on plasma and erythrocyte fatty acid composition and biosynthesis of prostanoids . Am J Clin Nutr 1987; 46:570–6.

  28. 28. Bjørneboe GEA, Johnsen J, Bjørneboe A, Bache-Wiig JE, Mørland J, Drevon CA. . Diminished serum concentration of vitamin E in alcoholics . Ann Nutr Metab 1988; 32:56–61.

  29. 29. SAS Institute Inc. SAS/STAT guide for personal computers, version 6 edition. Cary, N.C.: SAS Institute, 1987.

  30. 30. Hamazaki T, Urakaze M, Makuta M, et al. . Intake of different eicosapentaenoic acid-containing lipids and fatty acid pattern of plasma lipids in the rats . Lipids 1987; 22:994–8.

  31. 31. el Boustani S, Colette C, Monnier L, Descomps B, Crastes de Paulet A, Mendy F. . Enteral absorption in man of eicosapentaenoic acid in different chemical forms . Lipids 1987; 22:711–4.

  32. 32. Lawson LD, Hughes BG. . Human absorption of fish oil fatty acids as triacylglycerols, free acids, or ethyl esters . Biochim Biophys Res Commun 1988; 152:328–35.

  33. 33. Fleischman AT, Bierenbaum ML, Stier A, Somol SH, Watson P, Naso AM. . Hypotensive effect of increased dietary linoleic acid in mildly hypertensive humans . J Med Soc N J 1979; 76:181–3.

  34. 34. Malhotra SL. . Dietary factors causing hypertension in India . Am J Clin Nutr 1979; 23:1353–63.

  35. 35. Margetts BM, Beilin LJ, Armstrong BK, et al. . Blood pressure and dietary polyunsaturated and saturated fats: a controlled trial . Clin Sci 1985; 69:165–75.

  36. 36. Leaf A, Weber PC. . Cardiovascular effects of n-3 fatty acids . N Engl J Med 1988; 318:549–57.

  37. 37. Fox PL, DiCorletto PE. . Fish oils inhibit endothelial cell production of platelet-derived growth factor-like protein . Science 1988; 241:453–6.

Citing Articles (312)

    Figures/Media

    1. Table 1. Composition of the Fish-Oil and Corn-Oil Supplements.
      Table 1. Composition of the Fish-Oil and Corn-Oil Supplements.
    2. Table 2. Characteristics of the Subjects with Hypertension in the Fish-Oil and Corn-Oil Groups.*
      Table 2. Characteristics of the Subjects with Hypertension in the Fish-Oil and Corn-Oil Groups.*
    3. Table 3. Intake of Nutrients and Body Weight before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*
      Table 3. Intake of Nutrients and Body Weight before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*
    4. Table 4. Plasma Phospholipid Fatty Acid Concentrations before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*
      Table 4. Plasma Phospholipid Fatty Acid Concentrations before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*
    5. Table 5. Comparison of Changes in Blood Pressure before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*
      Table 5. Comparison of Changes in Blood Pressure before and after 10 Weeks of Supplementation with Fish Oil or Corn Oil.*
    6. Figure 1. Changes in Mean Arterial Pressure during the Study, According to Group and Number of Fish Dishes per Week in the Usual Diet.
      Figure 1. Changes in Mean Arterial Pressure during the Study, According to Group and Number of Fish Dishes per Week in the Usual Diet.

      Each point represents the group mean adjusted for age, sex, smoking (yes or no), and changes in body weight from week 0 to week 10; standard errors of the mean ranged from 1.4 to 1.8. The differences between values at week 0 and the arithmetic means of the measurements at week 5 and week 10 were tested by analysis of variance.

      The change in blood pressure differed significantly (P<0.0001) between the subjects with a low intake of fish in the fish-oil group (■—■) and in the corn-oil group (■— —■), but not (P = 0.28) between the subjects with a high intake of fish in the fish-oil group (▲—▲) and in the corn-oil group (▲— —). The change in blood pressure differed significantly between the subjects with high and low intakes of fish within both the fish-oil (P = 0.0008) and the corn-oil (P = 0.048) groups.

    7. Figure 2. Changes in Mean Arterial Pressure during the Study, According to Group and Base-Line Plasma Phospholipid n—3 Fatty Acid Concentrations.
      Figure 2. Changes in Mean Arterial Pressure during the Study, According to Group and Base-Line Plasma Phospholipid n—3 Fatty Acid Concentrations.

      The subgroups are based on the median base-line plasma concentration of phospholipid n—3 fatty acids (the sum of alpha-linolenic, eicosatetraenoic, eicosapentaenoic, docosapentaenoic, and docosahexaenoic acid levels), which was 172.6 mg per liter in the corn-oil group and 175.1 mg per liter in the fish-oil group. Each point represents a group mean adjusted for age, sex, smoking (yes or no), and change in body weight from week 0 to week 10; standard errors of the mean ranged from 1.6 to 1.8. The differences between the value at week 0 and the arithmetic mean of the measurements at week 5 and week 10 were tested by analysis of variance. The values for two subjects in the fish-oil group and three in the corn-oil group were not available, and those subjects were not included in these analyses. Squares represent subjects below the group median, and triangles those above. Solid lines represent the fish-oil group, and broken lines the corn-oil group.

      The change in blood pressure differed significantly between the subjects below the group median in the fish-oil group and those below the group median in the corn-oil group (P = 0.0024), as well as between the two strata within the fish-oil group (P = 0.016).

    8. Figure 3. Changes in Mean Arterial Pressure during the Study, According to Changes in Plasma Phospholipid n—3 Fatty Acid Concentrations in the Fish-Oil Group.
      Figure 3. Changes in Mean Arterial Pressure during the Study, According to Changes in Plasma Phospholipid n—3 Fatty Acid Concentrations in the Fish-Oil Group.

      The values are the mean (+SEM) differences between base-line values and the arithmetic mean of measurements at week 5 and week 10, adjusted for age, sex, smoking (yes or no), and change in body weight (P = 0.027 for linear trend by analysis of variance). The values for two subjects were unavailable and those subjects were not included in these analyses. The changes in plasma phospholipid n—3 fatty acid levels were measured as the percent change (by quartiles) in the sum of alpha-linolenic, eicosatetraenoic, eicosapentaenoic, docosapentaenoic, and docosahexaenoic acids between base line and 10 weeks.

      More Research